Like blockchain, artificial intelligence (AI) is one of the biggest buzzwords in banking and very likely will reshape the way financial services are rendered today. But what exactly is AI and how can it be used in wealth management (WM)? What about its risks and can it replace human intelligence?

What is AI?

The term AI was coined in 1955 by the American computer scientist John McCarthy, based on the idea that “every aspect of learning or any other feature of intelligence can in principle be so precisely described that a machine can be made to simulate it”.¹

Other terms – like machine learning, smart automation, cognitive computing, self-service analytics – are all closely related to AI.

What all these terms have in common is that they imply a certain capacity to digest large volumes of complex, unstructured (real-time) data, which enables computers to read, write, speak, listen to, see and interpret that data.

To leverage this capacity, AI platforms need access to high-quality data in huge volumes and regular transactions.

The main drivers for the big data revolution we have already witnessed are volume, velocity and variety of data. Meanwhile, the value and validity of data still seem to be an issue across the WM industry as a whole.

Technology giants like Amazon, Apple, Facebook and Google have been using AI technologies for quite some time. In contrast, banks have been rather slow in adopting them, although there are more and more areas where AI is starting to generate real benefits:

• Marketing is a big thing already and a lot of firms – like BlackRock, Deutsche Bank, UBS or Wells Fargo – are using AI engines to analyse people’s digital footprints and behaviour, to predict which products and services they are most likely to want to buy.
• There is also a compelling case for RegTech solutions to improve risk management, helping banks to deal with know-your customer (KYC), regulatory reporting (e.g. under the Markets in Financial Instruments Directive 2 (MiFID2)), risk scoring, churning, etc. and using that information for marketing purposes.
• In addition, AI can help to reduce operational risks by improving (client) security and preventing fraud using biometric processes, like TouchID, face and iris recognition, or even heartbeats from wearables.

How Can AI Be Used in WM?

While these examples are about more efficient and effective processes, from a client’s perspective it is about better (i.e. faster, more objective and individual), cheaper (with lower costs resulting from automation) and convenient (anytime and anywhere) advice.

Robo-advisors are a good example of how digitization can help to enhance financial services. But looking at the main components of managing wealth, there are many more opportunities to make use of AI.

What About the Risks of Using AI?

Talking about AI and automated processes, there are genuine concerns about their systematic risks and over-reliance on computer models: algorithmic trading programs have caused serious volatility and flash crashes in the past, and the risk and pricing models invented by the quants in Wall Street in the early 2000s turned out to be based on subprime.

In order to prevent outcomes like this from happening, validating machine thinking by running endless tests and validation scenarios – as well as putting restrictions and stops in place – should be a top priority. However, this sometimes seems to mirror the debate about self-driving cars, in that the technology used in a number of driving-assist systems (and that is already helping to prevent accidents) is viewed as problematic, as long as automated cars are not 100% safe (which they will never be).

In order to address this, firms need to decide how and to what extent AI will be used, depending on the work in question: whether they simply use it to make better, faster and more informed decisions, use it to supervise work done by humans (or vice versa), or actually delegate decision-making to it.

Can AI Replace Human Intelligence?

Money, and what we do with it, is a highly emotional affair and sometimes completely irrational. In providing emotional support and dealing with feelings, advisors still have an advantage over machines.

At the same time, a lot of advisors are struggling to stay on top of the information needed to provide clients with good advice, or they spend too much time trying to obtain it. Using AI will actually allow them to reinvest some of that time and to focus on other aspects of their client relationships.

Another aspect of AI and automation is that they can be perceived as rather impersonal. But this is an argument one can easily turn on its head: using AI, individual investment proposals will be far more personal than many of the standardized solutions offered so far.

Hence, only those companies that can use the new technology in ways that resonate with the human desire for individuality and empathy, or use it to work alongside their client advisors, will be able to capture its full potential.

Conclusion

• AI will play a huge role in helping wealth managers to better communicate and articulate their value proposition, combining the best of customer relationship management (CRM), portfolio and risk management tools, in order to provide clients with better services and more substantiated advice.
• The capability to integrate and use AI will become a crucial factor in staying competitive over the next five years. At the same time, the new technology will attract new market participants and make some banks look more like IT providers, while others will use AI to become product innovators.
• Being part and parcel of a new service model, AI requires an extensive redesign of existing processes and needs to fit within the overall digitization and business strategy. It also requires a clear understanding of the product and design features and how it can be sourced.

If there is one thing to learn from what we are experiencing right now, it is that enhancing financial services technology is unlikely to be the limiting factor. Even in banking, data is the new currency.